Jared Miller

Results-driven Software Engineer with extensive experience in full-stack development, data annotation, and AI model training. Proven expertise in building scalable applications, optimizing machine learning workflows, and ensuring high-quality training data for AI systems. Strong background in software development lifecycle, data quality assurance, and collaborative team environments. Adept at bridging the gap between software engineering and AI/ML operations to deliver robust, intelligent solutions.

• AWS Certified Solutions Architect - Associate, Amazon Web Services, 2022
• TensorFlow Developer Certificate, Google, 2021
• Professional Scrum Master I (PSM I), Scrum.org, 2020

• Enhancing Neural Network Training Efficiency Through Adaptive Data Annotation Strategies, Undergraduate Thesis, University of Washington, 2017, Investigated the impact of dynamic data labeling techniques on deep learning model convergence rates, demonstrating a 28% reduction in training time while maintaining 95% accuracy. Proposed novel quality metrics for evaluating annotation consistency in large-scale computer vision datasets.
• Transfer Learning Applications in Multi-Domain Sentiment Analysis: A Comparative Study, Independent Research Paper, 2021, Analyzed the effectiveness of pre-trained transformer models across diverse text domains including social media, product reviews, and technical documentation. Achieved state-of-the-art performance with fine-tuned BERT variants, publishing findings in online AI research communities.

• AI-Powered Resume Analyzer, Python, TensorFlow, Flask, React, 2023, Developed an intelligent resume screening system using NLP and machine learning to match candidates with job requirements. Achieved 92% accuracy in skills extraction and job fit scoring through custom-trained BERT models. Deployed on AWS with automated CI/CD pipeline, processing 1000+ resumes daily.
• Automated Data Annotation Platform, Python, OpenCV, PostgreSQL, Docker, 2022, Built a web-based annotation tool for image and video labeling with collaborative features and version control. Integrated pre-labeling ML models to reduce manual annotation time by 40%. Implemented real-time quality metrics dashboard for tracking annotator performance and dataset statistics.
• Real-Time Sentiment Analysis API, Python, FastAPI, MongoDB, Redis, 2021, Created a high-performance API for analyzing customer feedback sentiment across multiple languages. Utilized transformer models (DistilBERT) and caching strategies to achieve sub-100ms response times. Processed over 5M text samples with 89% sentiment classification accuracy.